Develop

cclutils/_base.py

@@ -12,7 +12,10 @@ __all__ = [
     'read',
     'write',
     'read_from_directory',
+    'read_chunks_it',
+    'read_sentences_it',
     'get_tagset',
+    'sentence2str'
 ]
 
 
@@ -206,3 +209,30 @@ def read_sentences_it(filepath, tagset='nkjp'):
         yield sentence
 
     del reader
+
+
+def sentence2str(sentence, use_bases=False, tagset='nkjp'):
+    """ Return corpus2.Sentence as a string.
+
+    Args:
+        sentence: a sentence object (corpus2.Sentence).
+        use_bases: if set to True, the we take base forms
+            instead of taking the orths.
+
+    Returns:
+        a string representation of the input sentence object.
+
+    """
+    if isinstance(tagset, str):
+        tagset = corpus2.get_named_tagset(tagset)
+
+    text = []
+    for token in sentence.tokens():
+        text.append(" " if token.after_space() else "")
+        if not use_bases:
+            token_string = token.orth_utf8()
+        else:
+            token_string = token.get_preferred_lexeme(tagset).lemma_utf8()
+        text.append(token_string)
+
+    return "".join(text).strip()